Saboteur is a poor game that doesn’t do the little it does very well. As a bluffing game, it is pathetically one-note. As a take that card game, it is lifeless and undramatic. As a route building game, it is simplistic to the point of mindlessness. I know the game has earned its fair share of devotees over the 15 years since its release (there are some very enthusiastic reviews on BGG as evidence of this), but this is a view I do not share in the slightest. It’s a fairly unique title, sure, but it’s also shallow, repetitive, and obnoxious. In fact, the game’s flawed nature is so openly admitted to — even by its fans — that perhaps there’s not much of a reason for me to be writing this review. Thing is, though I agree with all the commonly acknowledged problems that Saboteur has, I personally find it to have many, many more.
Saboteur is a game about a group of dwarves mining for gold. Some of the players play as good little dwarves trying to get to the gold stash, while others play as the titular Saboteurs: evil, butthole dwarves eager to stop the gold from being found. The game is played over a series of three rounds. At the beginning of each the map of the mine is set up by placing the entrance card at one end of the table and three “Goal” cards face-down at the other, one of which is hiding the gold stash (the cave map ain’t as loosey-goosey as it sounds though, exactly 7 card-widths of space must be between the entrance and goals). Players are then dealt a secret role card that tells them whether or not they are a saboteur for the round and also a hand of player cards that includes different paths to add to the cave and various special actions.
A player does one of three things on their turn: 1) play a path card on the cave to extend the mining routes out from the start card, 2) play an action card and activate its ability, or 3) discard a card face-down. Regardless of which they choose, they end their turn by drawing a replacement card. Path cards come in a multitude of forms: left turns, right turns, crossroads, etc. Some of them even feature dead-ends, an exposed saboteur’s favorite. As far as action cards go, there are only a few. Three different kinds of “Broken Tool” cards that all do the same thing: prevent a player from playing path cards, three different kinds of “Fix Tool” cards for dealing with the corresponding broken tools, “Rockfall” cards for removing previously placed path cards, and “Map” cards which let you look at one of the goal cards to see if it is hiding the gold or not. That’s every card in the game. Like I said, the game is simple.
A round of Saboteur ends when either a path to the gold has been finished, or everyone runs out of cards and said path is still not completed. At that point, players reveal their roles (everyone will already know who the saboteurs are, so this is largely just a formality). If the path has been completed successfully, the player who played the final path card draws a number of gold scoring cards equal to the amount of non-saboteur players, each randomly worth 1-3 points. They select one and pass the rest clock-wise, where the next player selects one and so forth. However, if the saboteurs prevented the path from finishing they do not score this way and instead earn a set amount of points for the round. This scoring system is super weird for a number of reasons, chief among them being that everything about Saboteur‘s design makes it feel like a team vs. team game, but it’s actually not. Since it’s played over a series of rounds, and it is randomly determined each round who the saboteurs are, players actually score on an entirely individual basis. This creates a bizarre friction between the social and mechanical components of the game that is extremely unintuitive. There’s a reason Werewolf and The Resistance have us vs. them win/loss conditions and not arbitrary player-specific point allotments. When you’re on a team in a game, it is only natural to behave in a manner that benefits the team.
For an example of this at its worst, consider the following scenario: it’s the third round, you are a non-saboteur, and the other players whom you are certain are non-saboteurs are all a point or two ahead of you in scoring. What the game would have you do at this juncture — in order to maximize your personal chance of victory — is to impede your own team’s progress just the right amount that you get to be the player who lays the final path card and gets first crack at the gold scoring cards, but not so much that the path never gets finished at all and the saboteurs end up earning the final set of points. What an obnoxious, dispiriting situation to be in. Helping your enemies to maybe help yourself if you’re lucky. But what if the other players start thinking you’re a saboteur? Uh oh! They might break your tools, oh no! (There are Saboteur diehards that swear by an optional rule that non-saboteurs with broken tools don’t get to score at the end of a round, because it actually makes this a concern. I’ve never played this way, but I get why it would be an improvement.) Regardless of how this all plays out, it feels mostly meaningless. You have no idea what cards any of the other players have, so there’s not a heck of a lot of information to go on even if you wanted to navigate this tightrope. It’s not interesting at all. Nothing in Saboteur is interesting. Semi-co-operative systems can be intriguing and thematic if implemented correctly, but the tools simply aren’t here for players to steer themselves through such a subtle social modality with any sort of nuance or intelligence.
Another problem with the game is that everyone’s turns for the first few trips around the table are more or less pointless. The main reason for this is that there is no way for a saboteur to meaningfully interfere with the progress of the table without making it immediately obvious who they are, so the sole source of tension in the game doesn’t even exist for the opening moves of a round. The saboteur can’t play dead-ends, obviously unhelpful path cards, or break anyone’s tools or else they will instantly become a target. They can play a map and lie about what’s under the goal card they look at, but they’re bound to get found out when a few others play their own maps and look for themselves. Maybe they’ll get lucky and the maps will all be in the hands of other saboteurs, but that’s hardly a satisfying result of their own cleverness. Even if they successfully deceive the table, the goal cards are all in the same general direction, so it doesn’t really matter all that much if players start digging toward the wrong one. Basically, all a saboteur can do early on is discard good path cards and wait for a time to strike. Once that happens, the cards that happen to be in the players’ hands define the rest of the round. I’m not even fully convinced that it matters if the saboteurs bother trying to hide their identities, because I’ve played a round where the saboteurs made themselves known on their first turns and still ended up winning the round because the order the cards ended up coming out. Pretty weak for a game that is primarily categorized as a bluffing game!
Saboteur is a soggy mush of basic systems that nevertheless constantly interfere with each other. Individual player decisions have very little weight due to the massive amount of hidden information and the chaotic back and forth of the action cards. This wouldn’t be so bad if it were a team game, because success would then largely be determined by its social component. But it’s not a team game, so its social component ends up being pretty much irrelevant. A player can, and at least one probably will, be on the winning side all three rounds and still lose. This is a game where what you should do on your turn is almost always obvious, but whether doing it leads to success or not lies entirely outside of your control. Yeah, it’s not so bad as the games where it doesn’t matter what you do at all, but easy choices that barely affect your chances of victory are only marginally better. As such, I highly doubt it is possible to be “good” at Saboteur, and if it is, I don’t care enough to find out how.
Saboteur gets a rating of TWO out of FIVE, indicating it is NOT RECOMMENDED.